• No results found

Europe information highways. Which standards? Research and Documentation Papers, Economic Series, Working Papers W-18, 2-1995

N/A
N/A
Protected

Academic year: 2019

Share "Europe information highways. Which standards? Research and Documentation Papers, Economic Series, Working Papers W-18, 2-1995"

Copied!
70
0
0

Loading.... (view fulltext now)

Full text

(1)

EUROPEAN PARLIAMENT

Directorate General for Researc_h

WORKING PAPERS

EUROPEAN INFORMATION

HIGHWAYS

Wmcn.

STANDARDS

?

E

c o

n

o

111

i

c

S

e r

i

e s

(2)

, This publication is available in French (original) and English (translation). Summaries included in the 11 officiallang}:lages.

The opinions contained in this document are those of the author and not in any way those of the European Parliament.

Reproduction and translation permitted except for commercial purposes and subject to the following conditions: the somce must be mentioned, and the publisher must be informed beforehand and provided with a copy.

Publisher: European Parliament_ Directorate-General for Reseaneh L-2929 Luxembourg

Author: Cecil AMElL

Official responsible: Anton LENSEN;

Internal Market Division

tel: (352) 43001

fax: (352) 434071

X400: S=LENSEN; «-'=FR; ADMD~ATLAS;

PRMD = PE WJ?M

(3)

'For all their newfound enthusiasmfor the Infobahn, unless something radical changes in the next few years, European-based companies aren't going to be riding the

Information Highway:

they'll be part of the pavement - run over in their own

backyards by companies with American and Japanese license plates. An era of European techno-sclerosis could follow, with

serious long-term effects on Europe's economies. '1

'Is there really no harm in dwelling on European backwardness when the Old Continent is oveiflowing with universally acknowledged

achievements, projects and expertise? ( .. )

By allowing the other side to choose the playing field, the European Union is obliged to leave everything to their initiative, to move forward at the pace they impose,

to play on unequal terms and, ultimately, to retreat on other fronts. 12

1 in Newsweek, 31 October 1994, 'Lost on the Infobahn?'

2

(4)

Contents

Foreword . . . v

S ummary . . . Vtl .. Avant-propos {in French) . . . . . . . . . . . . . . . . . . . . . . . ix

Resume {in French) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Index . . . 47

Summaries in other official languages . . . 49

Introduction . . . 1

• Is there a question mark against international standardization? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 3

• In what way are standards essential for information technolo~ '? . . . E> • What is the reason for the success of the Internet ? . . . . . . . . . . . . . . . 11

• What are the consequences of this success? . . . 13

• What are the shortcomings of ISO ? . : . . . . . . . . . . . . . . 1 E> • What is the European Union's policy on the subject ? . . . . . . . 19

• Are these different objectives compatible'? . . . 21

• Complementarity or competition between the two systems of standardization, OSI and IPS ? . . . 23

(5)

• What stage have the Europeans reachecdi ? . . . • . . . . 30

• What would be the consequences of int!ernational

recognition of IPS ? . . . .. _ . . . . 34

• The question of information content ('€mntentware') . . . 40

• Is there still time to reverse the observed· trends or should European strategy simply be to adapt tl(!)) the

reality of the Internet ? . . . 43

• What aims for Europe ? . . . .. . . . .. 45

Further iriformation

B" The OSI standards . . . .. .. . . . 8

(6)

Preface

One of the central priorities of the European Union is to develop a common European information area which will be embedded in the global information infrastructure. It is in the interest of Europe, as the world's largest trading bloc, with its diversity of cultures and its pluralism, to have an information infrastructure open to the world.

The Commission's vision of such an infrastructure was first presented in the December 1993 White Paper on 'Growth, Competitiveness and Employment'. 1 That vision was later taken up in the Bangemann Report on 'Europe and the Global Information Society',2 presented to the European Council of Corfu in

June 1994, and in the Action Plan3 issued by the Commission in July 1994.

The role of Information Technology (IT) standards is one of the least understood and least publicized items in the discussion about the information society. Although these standards are of a technical specialist nature, their choice and effects are highly political. They determine how we will communicate electronically in Europe and globally; who will communicate and at what cost (universal access, freedom to communicate); the European and national procurement policies for IT equipment, software and services; and the competitiveness of most sectors of economic activity dealing with information as a strategic resource.

Standards offer the guarantee of global electronic systems interconnection, to the benefit of the media that are used as a powerful means of communication and a tool of international trade.

The present report analyses the political issues surrounding global information network standards, i.e. the strategic choice between standards and protocols developed in the framework of international and European standardization or de

facto market-tested standards in the Internet.

1

Available on World-Wide-Web: http://www.echo.lu/en/wpaper/contents.html.

2 http://www .eam.net/EC/bangemann.html.

(7)

This choice is imminent because the progress in technology combined with liberalization of information infrastructures has provided critical mass for the explosion of global multi-media information services.

Since the technologies of data -processing, telecommunications and broadcasting media are converging, IT standardization policies not only determine how information content is transmitted, or how open or interconnected networks are. These policies are also inextricably associated with issues like the suitability of standards for operation in a multilingual environment, and the control and security of information and data privacy.

IT standardization therefore touches upon the very sensitive areas of cultural identity, cultural diversity and national security.

(8)

Summary

Since the American Administration dropped its plans to provide exclusive public support for the international standardization process in IT (Information Technologies), in favour of standards used for the Internet, the future of OSI (Open Systems Interconnection) standards has become highly uncertain. The consequences for the rest of the world could be particularly great in view of the size and commercial strength of the American IT market.

Europeans have good reason to believe that this decision has called into question their own standardization procedure, which closely follows that used for OSI.

Imminent international recognition of the IPS (Internet Protocol Suite) is likely to force a new look at European legislation on· public procurement. In addition, it will bring out into the open competition between two sets of standards which are not, in theory, on the same level. An assessment of their different objectives and possible (in)compatibilities does not appear to indicate that the Open Systems philosophy should be abandoned in the European Union. However, the commercial reasons underlying the rapid expansion of the IPS and, in particular, the formidable support from which the latter has benefited over many years thanks to the success of the Internet and applications on the World-Wide-Web leave little doubt as to the outcome of this forced co-existence.

'

From a comparison of the history of the Internet and the progressive development of the OSI, it must be admitted that the European long-term guidelines which have been predominant hitherto seemed to suit the particular political and cultural character of the Old Continent. The proliferation of European 'para-standardization' groups and bodies, the promotion of standards by public procurement, and the moves to overcome the slowness of the international negotiations bear witness to the determination with which European standardization policy has been implemented. The constant concern of the European Commission has been maximum involvement of as many players as possible in an open and consensual process, in accordance with the spirit of OSI.

(9)

These issues are not only urgent; they raise fundamental questions concerning Europe's technological and socio-cultural future, especially in the light of the

information highways. The choice of standards in these sectors will have a major impact on the technology and types. of information distributed on the electronic networks.

If US trade policy is conceded advantages in an area as strategic as the information industry, market forces could well become distorted. Furthermore, is the market the only mechanism to be relied on when it comes to responding to the particular needs of Europe with regard to 'contentware' and information infrastructure?

It is evident that standards in the area of infermation are so important that they warrant a definite decision in favour of:

a decidedly more affrrmative commercial strategy on the part of the European Union, in order to liberate market forces which are the principal motivator of innovation and finance - as· the Bangemann report proposes; a simultaneous reflection on the most appropriate standards for Europe's specific characteristics: despite the commercial pressures, it is essential to avoid hasty decisions, which may be irreversible, regarding the future implied by Europe's entry into the 'new age of the techno-industrial system'.1

1

in l&T Magazine, Spring 1994, 'Transeuropean Information Networks', G. Santucci, DG XIII of the

(10)

AUTOROUTES EUROPEENNES DE L'INFORMATION VERS QUELLES NORMES?

Avant-propos

L'une des principales priorites de !'Union europeenne est de developper un espace europeen commun de !'information, construit a partir d'une infrastructure globale de !'information. C'est dans !'interet de !'Europe, premiere puissance commerciale de la planete, riche d'une grande diversite de cultures et soucieuse de preserver son pluralisme, de disposer d'une infrastructure de !'information mondialement ouverte.

La perception que la Commission europeenne a d'un tel ensemble fut presentee pour la premiere fois en decembre 1993, dans le Livre blanc sur 'la croissance, la competitivite et l'emploi'.1 Cette vision fut ensuite reprise dans le rapport sur

'L'Europe et la societe de !'information planetaire'2 du Groupe de haut niveau

sur la societe de !'information, place sous l'egide du commissaire europeen M. Bangemann et presente au Sommet de Corfou enjuin 1994, ainsi que dans le Plan d'action3 de la Commission 'Vers la societe de !'information en Europe',

presente en juillet.

Le role des normes dans les technologies de !'information [TI) figure parmi les aspects les plus meconnus et les moins reveles dans les de bats sur la societe de I 'information.

Bien que ces normes soient de nature technique et relevent de specialistes en la matiere, leurs choix et leurs effets sont hautement politiques. Elles determinent en effet la maniere dont seront assurees les communications electroniques, en Europe et dans le monde; l'etendue du public en mesure de communiquer, eta quel cm1t [service universe!, liberte de communiquer); les politiques nationales et europeennes concernant les marches publics des equipements, services et logiciels des TI; et la competitivite de la plupart des secteurs de l'activite economique qui traitent de !'information comme d'une ressource strategique.

1 Disponible sur le World-Wide-Web: http://www.echo.lu/en/wpaper/contents.html.

2 HTTP://www. eam.net/EC/bangemann.html.

(11)

Les normes offrent !'assurance d'une connexion des differents systemes electroniques sur !'ensemble de la planete, au benefice des differents media qui sont utilises comme un instrument puissant de communication et un outil du commerce international.

Le present rapport analyse les enjeux politiques relatifs aux. normes utilisees pour les reseaux d'information de la planete, c'est

a

dire le choix strategique

a

effectuer entre, d'un cote, les normes et protocoles developpes dans le cadre institutlonnel international et europeen et, d'un autre cote, les standards de

facto, promus par le marche et utilises pour Internet.

Ce choix doit etre imminent, carle progres technologique combine aux. effets de la liberalisation des infrastructures d'information a deja permis d'atteindre une masse critique favorable

a

Ia floraison de services globaux d'information multimedia.

Etant donnee la convergence des technologies qui assurent le traitement de !'information [informatique), les telecommunications et la radiodiffusion, les politlques de normalisation dans le domaine des TI ne determinent pas seulement la maniere dont !'information est transmise ou les reseaux interconnectes. Ces politiques doivent aussi, inevitablement, s'attacher

a

des problemes telles que l'aptitude des normes

a

operer dans un environnement multiculturel et multilingue, ou le controle, la securite de !'information et la protection des donnees personnelles.

C'est la raison pour laquelle, dans le secteur des TI, les normes sont inextricablement liees aux. questions tres sensibles de l'identite culturelle, de la diversite culturelle, de la securite nationale.

(12)

Resume

Depuis que le gouvernement federal americain a envisage de ne plus apporter un soutien public exclusif au processus international de normalisation dans les TI (technologies de !'information), au profit des standards utilises par Internet, l'avenir des normes OSI. (Open Systems Interconnexion) est devenu particulierement incertain. Les consequences pourraient etre tres importantes pour le reste du monde compte tenu de la taille et du pouvoir d'influence commerciale du marche americain des TI.

Les Europeens ont de bonnes raisons de croire qu'une telle decision remet fondamentalement en cause leur propre strategie normative, qui s'inspire etroitement de celle adoptee pour l'OSI.

La reconnaissance internationale du protocole Internet IPS (Internet Protocol Suite), sur le point d'aboutir, risque de compromettre la politique europeenne des marches publics. Elle va en outre officialiser la mise en concurrence de deux ensembles normatifs qui ne sont, theoriquement, pas sur le meme plan. Une evaluation des differents objectifs et (in)compatibilites possibles entre ces deux ensembles ne parait pas justifier !'abandon de la philosophie des Systemes Ouverts. Pourtant, la politique commerciale qui sous-tend l'essor de l'IPS et, surtout, le soutien formidable dont celui-ci beneficie depuis de nombreuses annees grace au developpement du reseau Internet et des applications electroniques sur le World-Wide-Web ne laissent pas vraiment place au doute quanta l'issue de cette coexistence forcee.

En considerant parallelement l'historique d'Internet et l'eclosion progressive de l'OSI, force est de constater que les orientations europeennes de long terme, qui ont prevalujusqu'a recemment, semblaient convenir tout particulierement aux specificites politiques et culturelles du Vieux Continent. L'essaimage de groupes et organismes europeens de paranormalisation, la promotion des normes par les marches publics, les initiatives pour surmonter les lenteurs du processus international montrent bien la determination avec laquelle 1a politique europeenne de normalisation a ete menee. La Communaute europeenne a toujours eu pour souci d'associer aussi etroitement que possible un maximum d'acteurs dans une demarche transparente et consensuelle, conformement a !'esprit prevalant pour l'OSI.

(13)

Ces interrogations non seulement revetent un caractere d'urgence, mais elles soulevent des interrogations essentielles sur l'avenir technologique et socio-culturel de !'Europe, notamment dans la perspective des autoroutes de

l 'information. Le choix des normes dans les secteurs en question conditionne en grande partie la technique et le type d'information dispensee sur les reseaux electroniques.

Si la politique commerciale americaine se voit conceder des avantages dans un domaine aussi strategique que I' information, lejeu du marche pourrait bien etre fausse. Par ailleurs, peut-on seulement s'en remettre aux. mecanismes du marche pour repondre aux. attentes des citoyens europeens, lorsqu'on songe aux. exigences en matiere de contenu de }'information et

a

la fmalite des infrastructures electroniques?

On le voit, les normes dans le do maine de I 'information sont a ce point determinantes qu'elles meritent des choix politiques fermes quanta:

!'adoption d'une strategie commerciale nettement plus offensive de la part de l'Union europeenne afm de liberer les forces du marche, moteur principal de I' innovation et du fmancement- comme y invite le 'Rapport Bangemann'; simultanement, une reflexion approfondie sur les normes les plus appropriees aux. specificites de !'Europe: en depit des pressions commerciales, il est essentiel d'eviter les decisions hatives, et peut-etre irreversibles, sur l'avenir que nous reserve I' entree dans un 'nouvel age du systeme technico-industriel'.1

1 in I&T Magazine du printemps 1994, 'Les reseaux d'information transeuropeens', G. Santucci, DG Xlli de

(14)

INTRODUCTION

INTRODUCTION

Information technologies (IT) cover a very wide field both in terms of issues and players.

It is therefore correct to say that universal and public service considerations are the main factors in:

• coverage of the whole territory under consideration so that all regions can take part in the information society;

• an appropriate response to the citizens' varying requirements; • high quality services at reasonable prices.

The appropriate physical support for this-major development and these characteristics is the infrastructure. -It has to be able to evolve to incorporate new technologies and new requirements. Because of the way in which they are promoted by national public procurement, it is the standards that determine the technology required for the realization of information infrastructures, as regards both networks and applications.

Even though the standards are complex technical documents, they represent first and foremost the outcome of political choices. So although the

standards may be developed in a spirit of voluntary cooperation, there is a likelihood that they will develop into instruments of competitiveness for a country or group of countries and even be used to break down markets such as the IT market into segments. This applies not only to proprietary

standards but also to standards in the public domain, particularly those developed within the context of an emergent national industry, national defence or, more generally, a protected sector. Once these standards have been tested they can be exported to, and marketed in, other sectors of the economy.

Recent events would seem to suggest that the US Department of Commerce may have adopted this political line.

When all is said and done, the information field has become extremely strategic: do the media not act as the standard-bearers of the values of the society they represent?

(15)

EUROPEAN INFORMATION HIGHWAYS: WHICH STANDARDS?

Information should therefore be looked at from two angles, economic and cultural. The present heated debates on the revision of the 'Television without Frontiers' directive is sufficient evidence of this.

In these circumstances, standards in the IT field concern aspects related to the container of the information and the content. They are crucially

important for the completion of the European Single Market and also for the

iriformation highways. These two factors are moreover inseparable: a European market implies the free movement of information and the creation of an open market for goods and services connected with information techniques.

The ongoing liberalization of telecommunications in Europe (infrastructure, services, equipment) is overshadowed by the fear in certain European circles that this opening up of the market will lead to the European market being flooded by information from across the Atlantic. Europe could once again fmd itself in a situation analogous to the prevailing situation in the

audiovisual world where fmancial and organizational capacity have become insufficient to ensure its independence.

(16)

A QUESTION MARK'/

Is there a question mark against international

standardization?

On the initiative of those responsible at national level for new technologies, the FIRP (Federal lnternetworking Requirements Panel), under the aegis of Diane Fountaine, Director of the Telecommunications Sector of the

Department of Defense of the American Government, published a report in April 1994 on the US GOSIP (Government OSI Profile), the federal public market for OSI products. 1

US GOSIP imposes internationally acknowledged specifications on all public calls for tender involving the telecommunications networks. In practice, the American federal agencies issuing calls for tender for telecommunications equipment and services are obliged to purchase OSI products. The

purchasing power of these agencies means that they thus influence the whole of the national market.

The European system of public tenders operates according to the same logic, under Decision 87/95/EEC of the Council of Ministers.2 The public sector

represents 15-20% of the European market in data-processing equipment and software. Conformity with the OSI standards presents four advantages:

it reduces public expenditure by preserving investment; it allows the

interconnection of different administrative networks; it acts as an incentive to their private partners to conform with the same type of standard; and it

creates a market.

It has, however, been ascertained that the American agencies have often bypassed the laws: they have avoided purchasing OSI products or they have discarded them in favour of products considered more suitable - but

duplicating the purchase.

1

In reality GOSIP stands for documents drawn up by several public administrations: UK GOSIP in Great Britain, US GOSIP in the United States; and the American organization responsible for disseminating standards applicable to the federal administration is NIST (National Institute of Standards and Technology). By a linguistic twist, US GOSIP has become the teem used to describe the federal public procurement market itself.

2 Council Decision of 22 December 1986 on standardization in the field of information technology and

(17)

EUROPEAN INFORMATION HIGHWAYS! WHICH STANDARDS?

The FIRP1 report therefore suggests a radical evolution of the US GOSIP

policy, in particular by:

• no longer giving mandates to federal agencies issuing calls for tender but only recommendations;

• authorizing the introduction of the IPS (Internet Protocol Suite), on the basis of which the TCP/IP (Transmission Control Protocol/Internet Protocol) standards have been developed in parallel with OSI (Open Systems Interconnection);

• giving the same recognition to IETF (Internet Engineering Task Force), the organization which prepares and draws up the IPS standards, 2 as to ISO (International Organization for Standardization);

• offering the possibility of using different protocols together in such a way as to be able to use several different types of standards on the same information infrastructure;

• allowing the federal agencies to acquire normative products from other international consortia than IS0,3 and even private proprietary

standards.

A four-year old project aimed at solving the problems of interworking

between the databases of the different departments of the Administration by using OSI standards has also been called into question.

The American aerospace industry, along with a large number of IT equipment suppliers and national governments throughout the world, reacted very strongly to these proposals, believing that they would ruin the long-term efforts to construct universal interworking via OSI. According to the Boeing information services, 'scrapping the enormous OS! investment

infavour of PAS", such as the ancient IPS; demonstrates a naivety in understanding the complexity of our modern world'. 5

The Department of Commerce nevertheless decided, in September 1994, to follow the recommendations made by FIRJ?l.

1 For a presentation of the content of this report and the· reactions to it, see the OSN review, Open Systems

Networking & Computing, Technology Appraisals Ud, February 1994 issue, and Reuters News Agency, 31 March

and 26 September 1994.

2 With the approval of IESG (Internet Engineering Steering Group) and backing from lAB (Internet

Architecture Board), two voluntary organizations like IETF.

3 e.g. XJOPEN, connected with the standardization of AT&T operation systems, and also ATM Forum.

4 'Publicly Available Specificities', standards developed on some regional markets.

"

(18)

A QUESTION MARK?

Such a U-tum may seem surprising. In 1986, the American federal Administration had clearly accepted GOSIP on the grounds that they could not unilaterally impose TCP/IP standards on other countries: these standards were mainly meant for their domestic market.

But eight years later, account has to be taken of the formidable success of the Internet and above all of the difficulties American frrms seem to have had with OSI -it being not complete enough to meet all their requirements and/or not profitable- which have made them prefer other standards.

The mandate in favour of US GOSIP has therefore been challenged on the grounds that OSI standards are not appropriate to users' needs:1 some commentators in the specialized press make no bones about talking of competition between TCP/IP standards and OSI standards and of the confrontation between two systems driven by fundamentally different mentalities with no hope of reconciliation.

The continuing incompleteness of the OSI standards is also emphasized, meaning that the overall objective of ensuring network interconnection (full-stack) has not been achieved; the few high-performance but isolated OSI standards2 would not stand up to comparison with a blueprint for an

iriformation highway such as the Internet.

1 'Currently there are no real advantages to moving to OSI. It is more complex and less mature than IP, and

hence doesn't work as efficiently'. Ed Krol, in The Whole Internet, User's Guide and Catalog, 1992, 'What Is the lnter1,1et?', 'What Does the Future Hold?', p.16.

2

(19)

EUROPEAN INFORMATION HIGHWAYS: WJUCH STANDARDS?

In what way are standards essential for

information technology?

'Once [technological/ products can be easily accessible to consumers. there will be more opportunities for expression of the multiplicity of cultures and languages in which Europe abounds'.1

The user of electronic infrastructures is looking above all for maximum user friendliness in access to information: he should not be aware of the existence of standards which remain in a way 'transparent'. It is clear that a suitable standardization policy will actively contribute to this accessibility.

Standardization is a determining factor for the interconnection of telecommunications networks and the interworking of services in information technologies. 'Interworking is giving everyone access to iriformation' according to Pascal Ozanne, head of Netware products with Novell. 2 Without high-performance standards, there would be no possibility of this.

The European Parliament's resolution on 'Europe and the global information society- Recommendations to the European Council' and on the

communication from the Commission entitled 'Europe's way to the

information society: an action plan' adopted on 30 November 19943 rightly

insists on the importance of standards for IT.

The text states that: 'the scope of the expected developments can at present only be guessed at ( ... ), which means that this unified regulatory

framework must be designedfrom the outset as an open-ended,

predictable and adaptable system'(§ K). Moreover, if the future information society is to operate on a global scale4 'it is therefore vital to coordinate the

efforts being undertaken by all the countries involved in the same process'

[§ L).

1

Europe and the Global/njonnation Society, recommendation to the European Council of 26 May 1994, known as the Bangemann report, p.16.

2

in Le Monde infonnati'que, 4 November 1994, 'L'introperabilite, pour quoi faire?'.

3

Parliament resolution on the information society of 30/11/94, based on report A4-0073/94 by Mr Femand Hennan, MEP.

4

(20)

ESSENTIAL IT STANDARDS

Consequently, 'it is incumbent upon the public authorities actively to

promote ( ... ) the swift establishment of, initially, European standards and, subsequently, international standards enabling networks and applications

to be interconnected { ... ) without any deterioration in performance or user-friendliness and encouraging the dissemination of multimedia products on

(21)

EUROPEAN INFORMATION HIGHWAYS: WHICH STANDARDS?

The OSI standards

1

'The OSI is a vast standardization process launched at the end of the 70s to meet a precise requirement, that of giving data-processing:systems the means of transferring iriformation and treating it cooperatively without. making any assumptions about the specific characteristics of their hardware or software. ll

In 1977, the International Organization for Standardization (ISO), created 30 years previously, alerted by the difficulties experienced in connecting data processing equipment of different brands and models, decided to launch a vast programme aimed at creating a network architecture, the OSI. The

Open Systems Interconnection (OSI), based on experience from

achievements with different kinds of networks, proposes standardization of external accessibility of computerized data, whatever the operating systems involved.3

There are two aspects of OSI standardization: the model, which is not a standard but a reference framework for the work of the standards setters; and a series of standards, defining the services and specifying the protocols as packages of rules governing communication between telecommunications equipment by way of stable interfaces (technical boundaries).

The fundamental principle underlying the· definition of OSI standards, which determines their essential characteristics as regards quality, conformity and interworking, is the breakdown of the overall logical structure of the model into functional modules (called layered architecture) for which the detailed specification can be entrusted to autonomous teams. Projects, tests and the finalization of modules can be conducted quite independently with the end result being virtually spontaneous integration - the modules remaining capable of subsequent modification.

1

Information taken from Concepts Reseaux -Concepts OSIIDSA, Une introduction aux systemes ouverts, Bull SA, Victor Chaptal de Chanteloup, January 1991.

2

in Concepts Reseaux -Concepts OSIIDSA, Une introduction aux systemes ouverts, op. cit, p.iii

3

Ideally, a distinction should be made between the area of Information Technologies (IT), i.e. informatics and its socioeconomic ramifications, and Telecommunications. Telecommunications comes under the standardizing activity of the ITU (International Telecommunications Union) and its specialized bodies (CCITT and CCIR) and, at European level, ETSI (European Telecommunications Standards Institute). OSI standardization essentially depends on the international bodies ISO and IEC (International Electronic Commission) and the European bodies CEN and CENELEC.

(22)

THE OSI STANDARDS

OSI architecture therefore incorporates a fundamental split between lower

layers, which provide a channel for the transmission of information and thereby have a transpartjunction, and upper layers, which use this function to transport the content of the communications between the application processes and in their tum provide a contact junction.

The low levels take as their reference the standards of the major networks, and in particular X25 (Transpac in France), which have existed since 1976 thanks to an active promotion policy on the part of the European

Community in favour of the interconnection and development of packet switching networks (EuroNet-Diane initiative). These major networks

generally use the communication resources offered by the public networks.

OSI standardization uses a worldwide decentralized structure managed by a centralized body, ISO. It is useful at this point to recall that before OSI, international communication between experts of different interests and geographical origins was extremely difficult, with each working session having to be prefaced by a long preamble in order to defme a common language.

It was only in 1982 that ISO started the process of endorsing preliminary drafts of OSI standards. And then it was not until the end of 1987 that a draft international standard was agreed. At the time, all networks suppliers were convinced of the necessity of being able to interconnect their systems. They expected the OSI to provide the means of managing on a global scale networks which were becoming more and more heterogeneous.

There are several types of official international standards, not only the OSI standards. But in the domain of information technologies, all of them are currently linked to OSI. Thus, for example, the International Electrotechnical Commission (IEC) and ISO work together in a technical committee, the JTCI; and in 1984 the CCITT (International Telegraph and Telephone Consultative Committee) and ISO agreed on the issuing of ISO standards and

recommendations on a common basis.

In the absence of specific binding legislation, conformity with these standards is not compulsory. 1 This is why a distinction has to be made

between three types of coexisting standards: proprietary standards (such as those of IBM in the 70s), the PAS (Public Available Specificities) and official standards. The latter may moreover be the result of official recognition of proprietary standards upgraded to PAS: one striking example is that of Windows, the Microsoft software which appeared in 1985 and which has become a universal reference in the world of micro-computers.

TCP/IP is an example of a non-officially recognized PAS at international level.

(23)

EUROPEAN INFORMATION HIGHWAYS: WmcH STANDARDS?

A clear distinction does however operate between de jure and de facto

(24)

The INTERNET'S SUCCESS

What is the reason for the success of the Internet

(International

Network of

Computers)?

'Although the Organizationfor International Standardization (ISO} was spending years designing the ultimate standard for computer networking, people could not wait. Internet developers, responding to market pressures, began to put their IP software on every conceivable type of computer. It became the only practical methodfor computers from different mamifacturers .to communicate'. 1

At the beginning of the 70s, in the United States as elsewhere in the

industrialized world, the practical problem was that of interconnecting large computers operating on closed systems: how could the different databases be interconnected?

The TCP/IP standards, which appeared at the end of the 60s, were developed in an American environment for American military and then scientific

networks as a means of transporting information from one network to another.

The Internet,

the network

of networks, which uses these standards, was developed from Arpanet (Advanced Research Project Agency NETwork), in the Pentagon (Department of National Defense). It was used by an

increasingly large body of people made up essentially of scientists,

universities and other academic users, usually simply for exchanging mail and data.

The situation now is that the Internet community represents a large

population of potential users characterized by the many registered addresses - the result of intensive use of electronic mail and file transfer on a network where these services are much less expensive and the public involved a very large one.

We have had to wait for about 25 years, however, for this communications architecture to become a social phenomenon. Having become the world's largest source of information accessible from a simple

micro-computer (Mac or Personal Computer) equipped with a modem, and also the

1

(25)

EUROPEAN INFORMATION HIGHWAYS: WHICH STANDARDS?

world's largest mailbox, the Internet makes it possible to exchange messages and documents in real time over distances of several thousands of miles 1

The public fmancial support it has been accorded, especially by the Federal Administration,2 has given it a decisive advantage: under the guise of public research investments, both in America and Europe, an infrastructure has been developed which offers connection possibilities by the leasing of

telecommunication lines, with access to all the information on the network, regardless of its origin or volume. This infrastructure has been rmanced by a global flat-rate system of charges which were paid into an independent

fund. 3 In other words the network was not free but the services were.

By way of comparison, OSI has always used an independent transport layer acting as a gateway to non-specific public networks, which presupposes a high utilization cost depending on the amount of information conveyed.

The use of the Internet has also been simplified by the fact that local computer networks (Local Area Networks, LAN) could connect up to the main Arpanet network and act as a relay for user groups;4 at the same time most of the Internet applications and software were available free on the network.5

It should also be mentioned that in the absence, for several years, of any other product, applications based on the TCP/IP standards became

established.

Apart from this last point, the main lesson to be learnt from this overview is that the Internet has in fact developed in a non-commercial

environment into the leading major international non-regulated network of added value services.

1 'From 1985 to April1994, the Internet has grown from about 200 networks to well over 30 000 and from

1 000 hosts (end-user computers) to over 2 million ( ... ) The traffic on the network is currently increasing at a rate of 6% a month', according to Jeffrey K. MacKie-Mason, University of Michigan (in FAQ (Frequently Asked Questions) 'Economics about the Internet' of 13 May 1994- available on the Internet network).

According to latest estimates (December 1994), the Internet covers 84 countries and some 32 million users.

2

The NSF (National Science Foundation) made a grant to support the principal network, officially in order for it to maintain its independence from the private sector.

3

A university in the United States pays an annual subscription of between 60 000 and 100 000 dollars for connection to a regional network of the Internet; in Europe, connection to a main network of 64 kilobits costs ECU 100 000 and to a high capacity network of 2 megabits, ECU 1 million.

4

The European institutions use the TCP/IP protocol for their local networks (LANs), together with the X25 protocol.

5 In 1979, Microsoft, which was to become the software Number 1, reached an agreement with AT&T for the

(26)

THE CONSEQUENCES

~

What are the consequences of this success?

'The Internet { ... ) this anarchic network has had the great merit of revealing, by way of a really large worldwide experiment, an immense need for services of a new type. This is why it would be dangerous for the Europeans to rfifuse .to have anything to do with this immense laboratory of interactivity'._l

This situation profited from more _or less active support from

dataprocessing professionals in Europe. Despite an official policy geared to developing OSI, some networks were even constructed on the basis of Internet protocols in the public, national and European sector.

At the present time, a growing number of voices are calling, even officially, for European policy to take full account of the PAS and to admit

international recognition of their existence - on the grounds that it is now essential to back those standards which are best adapted to existing markets. Recognizing the TCP/IP standards is thus a way of making Internet's European and world penetration official. .

The main supporters of PAS consider that it is only the questions of security of data and intellectual property which merit in-depth examination in order to establish the use of the relevant standards.

Despite the significance of TCP/IP standards it is nevertheless acknowledged that there is virtually no way of detecting the quality of the information sought on the Internet, and to classify, in terms of relevance, the flow of information available on the network. Moreover- and this is perhaps the most important aspect - the standards used on the Internet do not seem to suit the specific multicultural character of Europe. 2

However, despite saturation of the network and as yet limited

user-friendliness, the Internet is now seen as the ideal support for setting up information highways on the Old Continent. 3

A large n~ber of access points have already made an appearance in Europe: it is possible to consult the databases freely on the European

1

in Futuribles, Les enjeux du multimedia, October 1994: 'Les promesses de l'unimedia' by Xavier Dalloz and Andre-Yves Portnoff.

2

see section on the advantages of OSI, p.8

3 see Courrier International for the week of 6-12 October 1994, the editorial 'Internet, media du

(27)

EUROPEAN INFORMATION HIGHWAYS: WJDCH STANDARDS?

Union's ECHO (European Community Host Organisation) 1 server via the

Internet and the new 'I'm Europe' initiative of the Commission's DG XIII also uses the network.

A more revealing fact is that the most sophisticated information tool at present available on the Internet, the World-Wide-Web (WWW) was

conceived by ... Europeans, but is being operated by the Americans.2 Tim

Berners-Lee, of the CERN ('Centre europeen de recherche nucleaire', European Organization for Nuclear Research) in Geneva, is the principal developer of this concept for hypertextual searches of multimedia

documents. It is also interesting to note that the European Commission has proposed a contribution of some ECU 1.5 million for this project.

Why has the WWW not been operated on the basis of networks

corresponding to the OSI standards?3 Without doubt because it would not

then have benefited from the formidable development support offered by the Internet. The public success of software using the WWW is, moreover, one of the main reasons why the business world has taken such an interest in the Internet because this software promotes greater user-friendliness in the utilization of the network and offers the possible prospect of economies of scale in its operation.

The research organization RARE, itself subsidized by the Commission, openly supports the WWW which is seen as a key element in the

development of information services for researchers in Europe.

The Commission has also approved the creation of a consortium to develop an interconnectivity gateway using the Internet software W AIS (Wide Area Information Services) between CERN and MIT (Massachusetts Institute of Technology).

The Internet has therefore established itself as the world network.

Due to its success, the Internet has become an ideal support for the promotion of defacto standards intended for international

recognition.

The problem would not arise if this recognition was to be carried out in the framework of ISO. But certain industrial and political circles, not only in America, seem to think that the predominance of the Internet, a

homogenous network spanning the globe, practically means abandoning the

1 http://www.echo.lu/

2 In

the same way the development of the WWW servers and applications for the public (such as Mosaic or Netscape, which are called 'browsers' or 'killer software') is carried out principally in the USA.

3 The fundamental concepts of WWW constituted by the hypertext protocol (HTfP) and language (HTML) are

(28)

THE CONSEQUENCES

(29)

EUROPEAN INFORMATION HIGHWAYS: WWCH STANDARDS?

11?'

What are the shortcomings of ISO?

It would be simple to conclude that the dominance of IPS is due to the fact that it is more in line with market laws. At the same time it has to be

admitted that the development of ISO standards is particularly slow.

Whatever strategy is adopted, it is generally accepted that speedy

standardization is essential for the future of information

technologies (IT). The world system is subject to great pressure because it

has to take part in drawing up standards at regional level as quickly as possible in order to influence the fmal decisions.

European standardization, modelled on the world system, remains subject to these vicissitudes.

It should also be noted that:

o The time taken to develop an OSI standard is much greater than that required for TCP/IP standards. 1

o The documents providing information about the OSI standards are claimed to be sold by ISO at a high price and their availability on the Internet very limited; students and researchers with limited budgets are therefore not readily able to access them. 2

o The multilayered structure of OSI seems to be controversial for some people, 3 who believe that this makes the· standards too complex.

1

For example, it will have taken ten years to bring out the X400 (electronic mail) standard as compared with one year for the first version of the corresponding SMTP (Simple Mail Transfer Protocol) standard! It is true that the validation of two RFC (Request For Comments, addressed to the IETF) applications suffices for a new application on TCP/IP.

2

However, some OSI documents are available at a very reasonable price, and even free in some cases, from the International Telecommunications Union (ITU), IETF or ECMA (European Computer Manufacturers Association).

3 Avoiding the redundancies in the functions in the different layers is one of the official principles of the OSI

(30)

SHORTCOMINGS OF ISO

o Many people in the standardization field even believe that OSI technology 'is now obsolete. 1

We-should not however forget the historical context: the OSI applications are very recent. They only really started to become operational in 1992 whereas the frrst TCP/IP standards appeared in the 70s; twenty years difference! Moreover, the adoption procedure for ISO standards is very formal: it

attempts to find a formal basis from the widest possible consensus which does indeed take a long time.

The main point is that the development of OSI standard9o is not based on existing user practice on a sufficiently wide scale.

The OSI has suffered from bad promotion and this explains its lack of popularity. The Commission issued a warning in 1988 that 'the risk of regional differences [regarding universal standardization], though unwelcome, is insignificant compared with the risk that

if

OS! were to make a chaotic start, the situation would be dUfi.cult to rescue later, politically, commercially and technically'.2

The American Federal Aviation Administration believes that the slow development of OSI products is caused by the lack of a goverment political machinery to enforce the US GOSIP mandate and the willingness on the part of government engineers and public procurement officials to ignore this mandate. This phenomenon can also be seen in Europe.

Steve Kille, of the ISO DE consortium, admits, does he not, that 'US GOSIP has done enormous harm to OS!' because of an ill-adapted procedure?3

l, 'As far as the market is concerned, OSI now seems to be extremely passe, ( ... ) Everything progressed to Unix; but now Unix is becoming passe. Possibly, the most open definition of 'open' has yet to be achieved, but the initial open systems have become obsolete before they've become successful. Maybe openness, like beauty, is

in the eye of the beholder', according to Christopher Read, of Apertus Ltd, an England-based consultancy (in

CommunicationsWeek International, of 27 June 1994, 'TCPIIP: it's official').

2 'European standards' in Standardization_ in infonnation technologies and telecommunications, DG XIll of the

Commission of the European Communities, Fact Sheet, May 1988.

3 in CommunicationsWeek International of 14 November 1994, 'Rival E-mail camps forge uneasy pact'.

(31)

EUROPEAN INFORMATION HIGHWAYS: WHICH STANDARDS?

The commercial policy has been inadequate and even disregarded.

In reality, 'the users basically don't care whether X400 (electronic mail) uses X25 [transport networks} or not [in preference to TCP/IP networks]',

says Mike Simmonds, a British Petroleum consultant. 'But TCP/IP over a leased line is likely to be less expensive than X25.

If

X400 charges stay as high as they are, people will switch over to IP service providers'. 1

The ITU (International Telecommunications Union) itself, whih previously used only X400 electronic messaging, now provides all staff with full

Internet access. A range of electronic services were built around the Internet because it is the most widely accessible technology for data communications and less expensive than X400. 2

The slowness of IGOSS (Industry-Government Open Systems Specification) in combining the four North American OSI profiles (US GOSIP, EPRI,

MAP/TOP and COSAC) - a combination which should constitute a very

influential purchasing power for the adoption of OSI standards in this part of the world - has not favoured OSI either.

At all events:

• The success of the Internet is not circumstantial: it has benefited from fmancial and political support, something which is not controversial in

itself and may even be laudable, but which is undeniable. The regional TCP/IP standards and alternative technologies have had the full benefit of this.

• The American Administration no longer appears to want to give sufficient support to OSI, allowing competition to develop between standards of national origin and standards designed as the fruit of international cooperation.

The fact that the partisans of the Internet officially intend to endorse a situation which they have allowed to develop, by moving forward now to international recognition of IPS, at least has the merit of clarity. Thus, following a report on US GOSIP, an agreement was concluded in June

I 994 between OSI and IETF on some mutual recognition of Internet and ISO protocols and on the organization of joint activities.3

There exists therefore for the time being a formal liaison between the two organizations to accompany the ongoing process of IPS internationalization.

1

in CommunicationsWeek International of 14 November 1994, 'Rival E-mail camps forge uneasy pact'. It should be recalled that access to X400 via X25 requires an access payment which depends on the volume of information transmitted, whereas the payment for Internet is a flat-rate one!

2

in Communications Week International of 6 February 1994, 'Groups tap Internet to drive standards work'.

3

(32)

EU POLICY

~

What is the European Union's policy?

'It will identify measures for the improvement of planning and prioritization, for

facilitating consensus-building, for speeding up the standard-making process andfor

appropriate use of standards in the international context.' 1

Serious thought has been given to how to respond to the shortcomings of European standardization which is closely linked to international

standardization.

For this purpose, and following proposals contained in the Commission Communication 'Europe's way to the information society, an action plan', 2 a workshop was held in Brussels at the end of November 1994. Its

conclusions were as follows:

• proposal to set up quickly a high-level independent and industrial

strategic group to undertake a review of the situation regarding European IT standardization in preparation for the G7 meeting on 25 and 26

February 1995;

• extension of the terms of reference of ITSTC (Information Technologies Steering Committee) to make it into a European industrial organization responsible for taking account of market priorities in this field by

proposing draft standards as rapidly as possible; and reform of EWOS [European Workshop for Open Systems) into a cooperation structure more open to activities linked to standardization;

• adoption of PAS as officially recognized standards, but in a joint framework with the officially approved organizations;

• reorganization of the organizational structure responsible for European IT standardization through the creation of a single workshop;

• better utilization of R&D funds to make prototypes and development tools, available more rapidly.

To sum up, the policy of promoting standards through public

procurement seems to be jeopardized3 and the expectations of the

market have been made the central concern as regards standards.

1 in Europe's way to the information society, an action plan, Commission Communication COM (94) 347 final of 19/07/94, 'Standardization, Interconnection and Interworking', p.4.

2 COM (94) 347 final of 19/07/94.

(33)

EUROPEAN INFORMATION HIGHWAYS! WHICH STANDARDS?

In its plan of action set out in the May 1994 report, the group of prominent persons under Mr Bangemann also recommended a revision of European standardization procedures to bring them more into line with the market and thereby to accelerate the interconnection of networks and interworking of services and applications in the telecommunications sector.

Despite the priority still given to international standardization, this

(34)

CoMPATmiLITY OF OBJECTIVES

~

Are these different objectives compatible?

Underlying the defence of the TCP/IP standards, there is a dominant feeling that the priority now is to ensure the rapid development of protocols in

response to the needs of consumers throughout the world, essentially on the basis of short-term profitability.

On the one hand there is an increasingly clear move towards making the process of recognition of standards developed outside ISO official: Christian Huitema, chairman of the lAB (Internet Architecture Board), considers that the distinction between de

facto and

de jure

standards no longer holds water. 1 On the other hand it is hoped to improve the process of European standardization by aligning it more closely with the development of technologies available on the market.

It is therefore possible that we may see a reversal of priorities in the future with the strategy of some industrial circles strongly influencing international standardization policy.2

So not only is international standardization directly concerned, but

European policy, which takes its lead from that, could undergo a complete U-turn.

The authors of American commercial policy seem to be reticent about the idea of absorbing OSI into IPS. Following the report on US GOSIP the suggestion of dovetailing IETF better into the shaping of stable open standards - and thus avoiding incompatibility between OSI and TCP/IP

standards - was not adopted. Finally, a minimum of control and a maximum of decentralization in everything concerning the Internet has become a

matter of pride and a guarantee of the future commercial success of the network.3

1

in Le Monde infonnatique of 4 November 1994, 'Internet demain: plus sU.r et mieux commercialise'.

2

'[Concerning OSI and IPS,] the question isn't Europe versus the United States. The question is political

solutions versus engineering solutions', 'according to Marshall Rose, chairman of Network Management Working Group, IETF (in CommunicationsWeek, 22 November 1993, 'Is Internet for Europe?').

3 'The Internet is a loose amalgamation of computer networks run by many different organizations in over

seventy countries. (..) one of the advantages of the Internet is that it is so decentralized: information sources are located on thousands of different computers', according to Jeffrey K. MacKie-Mason, University of Michigan (in FAQ (Frequently Asked Questions) 'Economics about the Internet' of 13 May 1994- available on the Internet

(35)

EUROPEAN INFORMATION HIGHWAYS: WmcH STANDARDS?

The consequences of this situation, initiated by a country whose IT market Is the largest in the world and whose commercial influence Is decisive, could be fundamental.

If the European Union in its turn abandons its public policy of promoting products which conform with OSI, by giving up the lever of public

procurement, there will no longer be any serious competitor for IPS. However, most of the products which conform to TCP/IP standards have been marketed mainly by American frrms and Europe, where many

(36)

COMPLEMENTARITY OR COMPETITION?

Complementarity or competition between the two

systems of standardization?

'The ability of the Internet to speak OSI protocols should help the Internet to expand into

more countries ( ... ) Most" of Europe regards IP as a cultural threat akin to EuroDisney. Networks based on the OSI protocols are much more palatable for them.

lf the two protocols could coexist, everyone would be happy'. 1

Politically there are some who claim that we are witnessing

co-existen~e

between two systems with different underlying values and priorities.

On the one hand ISO/IEC, presented as a hierarchical control system: o public interest is a direct objective which presupposes positive action; o the public authorities should guide the choice of standards;

o there is a formal coordination structure, with a majority vote.

On the other, lAB and the associated bodies IETF and IESG, presented as

an autonomous decentralized group:

o public interest is the outcome of the confrontation of private interests; o the public authorities can only set major objectives, without imposing the

means for their achievement;

o the structure offers a more informal voting procedure and would be open· to all kinds of participants: the systematic search for a consensus depends on commercial issues.

This over-simplified comparison puts the emphasis on the supposed advantages of IPS over OSI, as a cumbersome, slow structure.

. '

At the same time the OSI philosophy is precisely to avoid the existence of a centralized system centred on a dominant constructor or group of

constructors. The declared aim is to develop the best tools for information network interconnection, of whatever geographical origin, purpose or type of equipment.

This explains the existence of a multiplicity of committees, technical groups and other organizations, and the numerous procedures for OSI

standardization. They make possible the application of the basic

standardization principles, i.e. transparency of decisions, possible participation by all parties concerned and the obtaining of the broadest possible consensus.

1

(37)

EUROPEAN INFORMATION HIGHWAYS: WHICH STANDARDS?

Apart from the technical question, are not guarantees for the development of stable standards using a participatory rather than centra~ed procedures exactly what industrial leaders are looking for?

In parallel with this, most of the technological decisions relating to the Internet are taken by small committees of volunteers1 which essentially ftx

the standards on the basis of market expectations. Despite a very

decentralized structure, some network functions, such as the standardization or attribution of electronic addresses, are taken care of by a centralized

'Internet Society'. There is no voting procedure nor is there any requirement concerning the application of these standards; the groupings of like-minded players and the technical solutions are worked out with a minimum of supervision.

In theory, at the end of the day, the comparison between, on the one hand a general open framework capable of assimilating de

facto

protocols and

making them into international standards, and on 'the other, a long-standing protocol which is in constant evolution, but is no more than a protocol, is

not relevant.

In practice, on the other hand, it has to be· admitted that the coexistence of the two systems persists in a situation~. of total confusion, to the

benefit of Internet applications.

The organization known as RARE {'Reseawr Associes pour la Recherche Europeenne', Associated Networks for EUFopean Research), subsidized by the Commission, promotes interconnectivft1y between European research networks by means of gateways between m5 et TCP/IP. 2 For this purpose RIPE ("Reseaux IP europeens", European lF networks) has been set up as the result of collaboration between European network operators using IPS.

Although it works openly for the combining of OSI and TCP/IP standards, RARE was the originator of COSINE (Cooperation for Open Systems

Interconnection Networking in Europe), wruch made it possible to create one of the firSt pan-European networks via IX.If (!Jntemational X25 Infrastructure) -replaced since February 1993 by EuropaNet.

Meanwhile, in the private sector, the EEMA (European Electronic Messaging Association) has recently authorized the dropping of the X25 public network in favour of IPS and is recommending more and more openly the

1

IETF, IESG and lAB (see Chapter 1).

2

(38)

COMPLEMENTARITY OR COMPETITION?

development of gateways between the standards for X400 (electronic mail) applications and TCP/IP standards. 1

As traffic from the Internet towards networks using X400 is six times greater than in the other direction and as for the present only one third of service suppliers on X400 offer access to the Intemet,2 it is easy to imagine the

outcome of this compromise.

So it is improbable that, in the years to come, the Internet and OSI will share the tasks without any problems, the first concentrating on hardware, and the other on software. It therefore seems optimistic to imagine that a

'Multiprotocol Working' will become the simple rule whereby a proportion of the applications using OSI standards will use Internet protocols.

The OSI standards developed for the Interlibrary Loan Protocol, for example, is likely to be used only as a mini urn basis of functional standards which are adapted to the real needs of the market, in parallel with the TCP/IP

standards. The few OSI applications which are geared to present requirements will thus be preferred to fundamental research and the anticipation of future requirements. 3

It is also planned that ISO should in the near future adopt the American ANSI/NISO Z39.50 standard.revised in 1992,4 which is used by a large

community of users on the other side of the Atlantic; whereas we have since 1991 had the SR (Bibliographic Search and Retrieval) standard, developed by ISO at the initiative of the US itself, with performance ratings comparable to those of the revised 239.50 ... This shows that an OSI standard can indeed anticipate market expectations.

1 The OSI/6000 standard makes it possible to use OSI applications to communicate with similar applications

on IPS networks, via RFC 1006 (especially for SMTP (Simple Mail Protocol) and FrP (File Transfer Protocol) applications) . EEMA is also proposing use of the RFC 1327 standard as an interconnection gateway between the Internet and X400.

2 According to Steve Kille, in CommunicationsWeek International of 14 November 1994.

3 In the library interconnection projects, the f'll'st concern is therefore the utility aspects of the OSI standards,

at the risk of disregarding the innovative aspects. This decision is based on the simple observation that most of the possibilities offered by the 10160/10161 OSI standards are not used (90% of the users of these two standards use only 50 per cent of the potential on offer).

4

An ISO meeting is scheduled for May 1995, in Canada, for this purpose.

(39)

EuROPEAN INFORMATION HIGHWAYS! WHICH STANDARDS?

As we can see, there is no lack of duplication between OSI and TCP/IP

standards. It is virtually a foregone conclusion that the market, supported by the Internet community, will have the fmal say.

Consequently, it is a legitimate exercise to look at the effects of the competition between the two systems, rather than improbable complementarity. 1

Of course there is at least one other example of non-conflictuallasting co-existence between an Anglo-Saxon system and a system originating in

continental Europe: the metric system. In this case the barrier is clearly a cultural one; but the economic implications and the consequences are clearly not the same.

The question then remains open as to whether the parallel development of the two systems of standards in the highly commercial strategic sector of IT

will not in the long run favour the disappearance of the OSI standards,

which although certainly very sophisticated are apparently less well-adapted to the American market and above all vulnerable to counter-arguments of immediate utility and cost.

1

'IP and X25 will be complementary, not exclusive, backbones', thinks Gilles Antoine, of the EEMA (European Electronic Messaging Association). 'Internet and X400 can coexist', believes John Mahoney, of lnfonet Software Solutions. Others consider that, on the contrary, SMTP will sweep aside X400 as it did X25 (in

References

Related documents

As with other rapidly reconfigurable devices, optically reconfigurable gate arrays (ORGAs) have been developed, which combine a holographic memory and an optically programmable

Turn on led_Red1 Turn on led_Green2 Wait 30 seconds Turn on led_Yellow1 Turn on led_Yellow2 Wait 3 seconds Turn on led_Red2 Turn on led_Green1 Wait 20 seconds Turn on led_Yellow1

the CaaX sequence) were appended to the C-terminus of the His 6 -EGFP-TEV proteins. Additional substrates were chosen based on high scoring peptides predicted to be FTase..

innovation in payment systems, in particular the infrastructure used to operate payment systems, in the interests of service-users 3.. to ensure that payment systems

Although all our member centers provide space for various 12 step meetings and other peer to peer recovery supports – recovery centers are not affiliated with any of these groups..

Marie Laure Suites (Self Catering) Self Catering 14 Mr. Richard Naya Mahe Belombre 2516591 info@marielauresuites.com 61 Metcalfe Villas Self Catering 6 Ms Loulou Metcalfe

The corona radiata consists of one or more layers of follicular cells that surround the zona pellucida, the polar body, and the secondary oocyte.. The corona radiata is dispersed

Figure 8 shows the evolution of the percentage of the population aged 30-34 years with tertiary degree in the southern countries of European Union on the total population